Date of Award

Spring 5-27-2025

Document Type

Thesis (Undergraduate)

Department

Computer Science

First Advisor

peter.chin@dartmouth.edu

Abstract

We propose that precisely timed neural activity cycles can serve as structural primitives for memory and computation in a system that exhibits associative learning like the brain. Inspired by biologically grounded mechanisms such as calcium-dependent plasticity, spike-timing-dependent learning, and phase-sensitive excitability, we construct a spiking neural network model in which repeated temporal coincidences drive the formation of self-sustaining activity loops. These cycles, once formed, persist as dynamic memory traces: not stored as static weights, but as reverberating patterns that replay in time when these loops are restarted. We show that noise alone fails to induce stable structure, but even sparse, structured input can trigger rapid cycle emergence, putting emphasis on sustained pattern of co-activation. These cycles are robust, yet remain flexible, selectively adapting to new patterns through localized phase-driven interference. Crucially, we find that memory formation, persistence, and forgetting arise from timing relationships alone, without supervision or global error correction. Our results suggest a paradigm in which rhythm, and not representation, underlies associative learning: it offers a biologically aligned alternative to gradient-based optimization and points to testable predictions about the role of phase-locking and temporal structure in neural systems.

Download

Share

COinS